Hepatitis-B virus infection predicts mortality of HIV and hepatitis C virus coinfected patients
Teira, Ramón; VACH Study Group
Correspondence to Ramón Teira, Hospital de Sierrallana, Calle Manuel Teira Fernandez s/n, 39300 Torrelavega, Spain. Tel: +00 349 4284 7400/67032; fax: +00 349 4284 7501; e-mail: firstname.lastname@example.org
Received 28 September, 2012
Revised 24 December, 2012
Accepted 9 January, 2013
To study hepatitis B virus (HBV)-hepatitis C virus (HCV)-HIV triple coinfection in the Spanish VACH cohort, we selected HCV-antibody positive patients who had recorded results for HBV surface antigen (HBsAg) and classified them as HBsAg positive or not. We compared their characteristics and outcome (death by any cause). Six thousand three hundred and seventy-nine patients fulfilled the inclusion criteria. Three hundred and fifty-five (5.6%) were HBsAg positive, which showed an association with mortality (mortality rate ratio: 1.90; 95% confidence interval: 1.42–2.54). This association persisted after adjusting for other prognostic variables.
There is considerable information about the coinfection with HIV and hepatitis C virus (HCV), as well as about coinfection with HIV and hepatitis B virus (HBV), as both have been common diagnoses in developed countries resulting from the highly efficient transmission of the three viruses through percutaneous injections, and from their ability to establish chronic infection.
Less attention has been paid to the coinfection by the three viruses Therefore, information regarding this condition, beyond the mere description of its prevalence in diverse settings, is scarce and inconclusive [1–5].
Here, we present a study for which we selected those patients included in the VACH Cohort, a large cohort of HIV-infected patients in Spain , who had a positive serologic test for HCV and a recorded result for HBV surface antigen (HBsAg). Patients were classified, according to the results of this test, as HIV-HCV infected and HBV coinfected (HBV+) or HIV-HCV infected without HBV (HBV−).
We selected ‘death from any cause’ as outcome variable, and Kaplan–Meier curves and Cox proportional-hazards models for the main analyses. Kaplan–Meier curves were constructed according to the classification by study groups, and also according to sex, ethnic group, mode of HIV transmission, AIDS at presentation, use of HCV treatment, use of antiretroviral treatment (ART), use of tenofovir and period (per 3 years) of diagnosis. We calculated follow-up from the date of the first visit to the VACH-associated clinic as the initial date until last visit to the clinic, predefined loss to follow-up, or death. We used Cox proportional hazards models to adjust for the effect of HBV coinfection on survival, controlling that of significant prognostic variables derived from the survival univariate analyses (see Kaplan–Maier above, and age, CD4+ cell count, plasma HIV-RNA concentration and AST-to-platelet ratio index; APRI), as well as sex and mode of HIV transmission. We categorized the use of ART into three exclusive groups: ‘never used ART’, ‘used ART but never tenofovir’ and ‘used ART with tenofovir’. We repeated the same analyses in a subset of patients for whom a test result of plasma HCV-RNA was available and positive.
Six thousand three hundred and forty-two patients fulfilled the inclusion criteria and contributed 25837.13 patient-years of follow-up. Of these patients, 355 (5.6%) were HBV+. Table 1 shows their main characteristics according to the category defined. Other variables studied, not shown in Table 1 and which were not significantly different between HBV+ and HBV− patients, were: time on ART, time not on ART, time after a first CD4+ cell count below 350, percentage of time on ART after a first CD4+ cell count below 350 and period of diagnosis.
We recorded 543 deaths [crude mortality rate: 2.10 per 100 patient-years (%PY), 95% confidence interval (CI): 1.93–2.28], liver-related death constituted the most common group of causes of death, accounting for up to 42.4% of those HBV+ with an identified cause and 23.6% in HBV−, followed by AIDS-related (24.2% and 22.3%, respectively), non-AIDS-related infections (6.1% and 14.6%), malignancies (6.1% and 14.6%), cardiovascular (6.1% and 10.4%) and respiratory diseases (3% and 0%). Hepatitis B coinfection was significantly associated with increased mortality (mortality rate in HBV+: 3.78% PY; 95% CI: 2.86–4.98; in HBV−: 2.01% PY, 95% CI: 1.86–2.22; mortality rate ratio: 1.90; 95% CI: 1.42–2.54). The following variables were also significantly associated with mortality in the univariate analysis: race/ethnic origin (P = 0.007), prior AIDS (P = 0.0001), treatment of hepatitis C (P < 0.0000), ever used ART (P < 0.000), age (P < 0.000), baseline CD4+ cell count (P < 0.000), APRI (P < 0.000) and log10 HIV-RNA plasma concentration (P < 0.000). Period of HIV diagnosis, sex and mode of HIV transmission were not associated. For the final Cox proportional-hazards model, we stratified according to ‘race/ethnic group’, which did not meet the proportional hazards assumptions, and forced in sex and mode of HIV transmission. The remaining variables presented above, except prior AIDS, remained associated with survival, with the following hazard ratios and 95% CI: age (per year older: 1.075; 1.055–1.095), CD4+ cell count (per 100 cells/μl increase: 0.877; 0.819–0.940), log10 HIV-RNA (1.154; 1.086–1.378), use of tenofovir-based ART (0.176; 0.109–0.286) and nontenofovir-based ART (0.352; 0.234–0.529), hepatitis C treatment (0.373; 0.165–0.843), HBV coinfection (1.745; 1.141–2.670) and APRI value (per unit 1.115; 1.045–1.191).
A test result for HCV-RNA was available for 3023 patients, of which 2716 (89.5%) were positive. HCV-RNA was detectable in 90.7% cases with a negative HBsAg and in 60.7% of those with a positive HBsAg test (P < 0.000). Survival was significantly worse among those HBV+. HBV coinfection remained associated with mortality in the Cox regression models, as did age and use of ART.
In a previous study of a subset of our cohort of HIV treatment-evaluable, HIV/HCV coinfected patients, we found that HBV infection was the strongest predictor for the development of end-stage liver disease . The hypothesis of a detrimental interaction between HBV and HCV is not new. In fact, studies exploring this issue were reported early after the identification of HCV. Overall, and in HIV-negative patients, HBV/HCV coinfection seems to be associated with a more severe liver disease than infection with HCV or HBV alone [8–13], the evidence being quite convincing for an association with hepatocellular carcinoma . Similar data were initially replicated in HIV-infected patients , although further studies provided inconsistent results [2–4]. In this setting, our results provide the strongest evidence to date that HBV infection significantly worsens the prognosis of HIV/HCV coinfected patients.
The mechanisms of the interaction between HCV and HBV, whether immunologically mediated, or at the cellular or molecular level, or a combination of any of them (or others), are only now beginning to be investigated. Initial attention was paid to the inhibitory effect on the peripheral expression of HBV attributed to HCV , but the opposite effect, of HBV on HCV, has been even better documented, in HIV infected [16–18] and noninfected [8,11,13] patients, but not uniformly [3,19]. Our results are consistent with the mainstream consideration on the occurrence of this effect.
The main strengths of our study are its longitudinal nature, long follow-up and large number of patients and of events. However, as in any observational study, the likelihood of unmeasured confusion is an important potential limitation that should be accounted for.
R.T. wrote the first draft and the final protocol of the study, received the database and performed the analyses, wrote the first draft and final version of the manuscript, and keeps the database and responsibility for the submission and validity of the manuscript. All of the remaining listed authors contributed equally to the discussion and design of the study, to the discussion of the results and to the writing and revision of the manuscript. All have read the submitted version and agree with it.
Conflicts of interest
R.T., Ignacio Suárez-Lozano, Fernando Lozano, Pere Domingo, Paloma Geijo, Bernardino Roca, Agustín Muñoz-Sanz, José López-Aldeguer, Enric Pedrol, Pepa Muñoz, Teresa Puig, Esteban Ribera and Vicente Estrada declare having received, at some time, travel grants, or speaking or advisory honoraria, from any of the following: Abbott Laboratories, Boehringer-Ingelheim, Bristol-Myers Squibb, Gilead, Glaxo-Smith-Kline, Hoffman-LaRoche, Janssen-Cilag, Merck Sharp and Dohme, Pfizer, and VIIV Healthcare. The remaining authors declare no potential conflicts of interests.
The writing group from the VACH Study Group: R.T. (Hospital de Sierrallana, Torrelavega), Ignacio Suarez-Lozano (Hospital Infanta Elena, Huelva), Fernando Lozano (Hospital de Valme, Sevilla), Pompeyo Viciana (Hospital Virgen del Rocío, Sevilla), Pere Domingo (Hospital Santa Creu i Sant Pau, Barcelona), Pepa Galindo (Hospital Clínico, Valencia), Paloma Geijo (Hospital Virgen de la Luz, Cuenca), Alberto Terrón (Hospital del SAS, Jérez de la Frontera), Jaime Cosín (Hospital Gregorio Marañón, Madrid), Esteban Ribera (Hospital Vall d’Hebrón, Barcelona), Bernardino Roca (Hospital General, Castellón), Maria L. Garcia-Alcalde (Hospital de Cabueñes, Gijón), Trinitario Sánchez (Hospital Virgen del Rosell, Cartagena), Agustín Muñoz-Sanz (Hospital Infanta Cristina, Badajoz), Alberto Romero (Hospital Clínico, Puerto Real), José López-Aldeguer (Hospital La Fe, Valencia), Enric Pedrol (Hospital Santa Tecla, Tarragona), Francesc Vidal (Hospital Universitari Joan XXIII, IISPV Universitat Rovira i Virgili, Tarragona), Myriam Garrido (VACH Data Management Center, Cartaya), Pepa Muñoz (Hospital de Basurto, Bilbao), Teresa Puig (Hospital Arnau de Vilanova, Lleida), Elisabeth Deig (Hospital General, Granollers), Vicente Estrada (Hospital Clínico de San Carlos, Madrid), Manuel Castaño (Hospital Carlos Haya, Málaga).
1. Ragni MV, Belle SH. Impact of human immunodeficiency virus infection on progression to end-stage liver disease in individuals with hemophilia and hepatitis C virus infection
. J Infect Dis
2. Castellares C, Barreiro P, Martin-Carbonero L, Labarga P, Vispo ME, Casado R, et al. Liver cirrosis in HIV-infected patients: prevalence, aetiology and clinical outcome
. J Viral Hepatitis
3. Laguno M, Larrousse M, Blanco JL, Leon A, Milinkovic A, Martinez-Rebozler M, et al. Prevalence and clinical relevance of occult hepatitis B in the fibrosis progression and antiviral response to IFN therapy in HIV-HCV-coinfected patients
. AIDS Res Hum Retroviruses
4. Bonacini M, Louis S, Bzowej N, Wohl AR. Survival in patients with HIV infection and viral hepatitis B or C: a cohort study
5. Sollima S, Caramma I, Mezanghi B, Massetto B, Acquaviva V, Giulani G, et al. Chronic coinfection with hepatitis B and hepatitis C viruses in an Italian population of HIV-infected patients
. J Acquir Immune Defic Syndrom
6. Suárez-Lozano I, Fajardo JM, Garrido M, Roca B, García-Alcalde ML, Geijo P, et al. Epidemiological trends in HIV infection in Spain: preventive plans have to be oriented to new target populations (Spanish VACH Cohort)
7. Teira R, Geijo P, Cosin J, Muñoz-Sanz A, Viciana P, Suarez-Lozano I, et al.Risk factors for end-stage liver disease among HIV and hepatitis C virus co-infected patients in the Spanish VACH Cohort
[abstract]. In: Abstracts of the Ninth International Congress on Drug Therapy in HIV Infection B
, Glasgow 9–13 November 2008. Abstract P134.
8. Zarski JP, Bohn B, Bastie A, Pawlotsky JM, Baud M, Bost-Bezeaux F, et al. Characteristics of patients with dual infection by hepatitis B and C viruses
. J Hepatol
9. Weltman MD, Brotodihardjo A, Crewe EB, Farrell GC, Bilous M, Grierson JM, et al. Coinfection with hepatitis B and C or B, C and delta viruses results in severe chronic liver diseases and responds poorly to interferon-alpha treatment
. J Viral Hepatol
10. Cacciola I, Pollicino T, Squadrito G, Cerenzia G, Orlando ML, Raimondo G. Occult hepatitis B virus infection in patients with chronic hepatitis C liver disease
. N Engl J Med
11. Crespo J, Lozano JL, de la Cruz F, Rodrigo L, Rodríguez M, San Miguel G, et al. Prevalence and significance of hepatitis C viremia in chronic active hepatitis B
. Am J Gastroenterol
12. Fong TL, Di Bisceglie AM, Waggoner JG, Banks SM, Hoofnagle JH. The significance of antibody to hepatitis C virus in patients with chronic hepatitis B
13. Mohamed AE, al Karawi MA, Mesa GA. Dual infection with hepatitis C and B viruses: clinical and histological study in Saudi patients
14. Donato F, Boffetta P, Puoti M. A meta-analysis of epidemiological studies on the combined effect of hepatitis B and C virus infections in causing hepatocellular carcinoma
. Int J Cancer
15. Liu CJ, Chen PJ, Chen DS. Dual chronic hepatitis B virus and hepatitis C virus infection
. Hepatol Int
16. Antonucci G, Vairo F, Iacombi F, Comandini UV, Solmone M, Piselli P, et al. Role of hepatitis B virus, hepatitis D virus and other determinants on suppression of hepatitis C viraemia in HIV infected patients with chronic HCV infection: A longitudinal evaluation
. Scand J Infect Dis
17. Soriano V, Mocroft A, Rockstroh J, Ledergerber B, Knysz B, Chaplinskas S, et al. Spontaneous viral clearance, viral load, and genotype distribution of hepatitis C virus (HCV) in HIV-infected patients with anti-HCV antibodies in Europe
. J Infect Dis
18. Martin-Carbonero L, Barreiro P, Jimenez-Galan G, Garcia-Berriguete R, Núñez M, Rios P, et al. Clearance of hepatitis C virus in HIV-infected patients with múltiple chronic viral hepatitis
. J Viral Hepat
19. Maida I, Rios MJ, Perez-Saleme L, Ramos B, Soriano V, Pegram PS, et al. Profile of patients triply infected with HIV and the hepatitis B and C viruses in the HAART era
. AIDS Res Hum Retroviruses
© 2013 Lippincott Williams & Wilkins, Inc.